No abstract

Automated vehicles (AVs) are a concrete possibility in the near future. As AVs may shift transportation paradigms, transportation agencies are eager to update their models to consider them in planning. In this context, the use of advanced models may be challenging, given the uncertainty in the use and deployment of AVs. In this paper, we present a general framework to extend the four-step model to include AVs, and test our extension on North Central Texas Council of Governments’ model. Our approach introduces a module for AV ownership and exogenous parameters into an existing four-step model to account for changes in travel decisions for AV owners, and for the impacts of AVs on network performance. The latter is modeled using the concept of passenger-car-equivalent to avoid imposing network-wide assumptions on AVs’ capacity consumption. We analyze five scenarios, representing different assumptions on the impacts of AVs, and include references to inform the selection of modeling parameters. We compute aggregate metrics that suggest that the model is sensitive to the proposed parameters, with the passenger-car-equivalent assumptions having the largest impact on model outcomes. Results suggest that, even when we assume that AVs can better use network capacity and that trip-making rates do not drastically increase, AVs may lead to an increase of about 2.8% in vehicle-hours traveled while also improving speeds by about 1.8%. If AVs introduce additional friction on traffic, the system performance may deteriorate. The analyses presented here suggest that existing modeling tools may be adjusted to support analyses of a future with AVs.

No abstract

No abstract

The introduction of mobile application-based ride hailing services represents a convergence between technologies, supply of vehicles, and demand in near real time. There is growing interest in quantifying the demand for such services from regulatory, operational, and system evaluation perspectives. Several studies model the decision to adopt ride hailing and the extent of the use of ride hailing, either separately or by bundling them into a single choice dimension, disregarding potential endogeneity between these decisions. Unlike developed countries, the literature is sparser for ride hailing in developing countries, where the demand may differ considerably because of differences in vehicle ownership, and availability and patronage of many transit and intermediate public transport (IPT) modes (the shared modes carrying 40% shares in some cases). This study aims to bridge these gaps in the literature by investigating three interrelated choice dimensions among workers in Chennai city: consideration of IPT modes, the adoption of ride hailing services and the subsequent usage intensity of ride hailing services. The main factors influencing these decisions are identified by estimating a trivariate probit model. The results indicate that sociodemographic and locational characteristics and the availability of IPT modes influence ride hailing adoption, whereas work-related constraints and perception of other modes affect its frequency. Work and non-work characteristics affect both the dimensions of ride hailing. Further, endogeneity is observed between ride hailing and IPT adoption after controlling for these variables, whereas evidence of endogeneity is absent among other dimensions. Mainly, the model separates the effect of the exogenous influences on the usage frequency level from their effect on the adoption of ride hailing services.

In consumer surveys, more information per response regarding preferences of alternatives may be obtained if individuals are asked to rank alternatives instead of being asked to select only the most-preferred alternative. However, the latter method continues to be the common method of preference elicitation. This is because of the belief that ranking of alternatives is cognitively burdensome. In addition, the limited research on modeling ranking data has been based on the rank ordered logit (ROL) model. In this paper, we show that a rank ordered probit (ROP) model can better utilize ranking data information, and that the prevalent view of ranking data as not being reliable (because of the attenuation of model coefficients with rank depth) may be traced to the use of a misspecified ROL model rather than to any cognitive burden considerations.

No abstract

This test paper develops and tests 13 direct ridership models (DRMs) for transit sketch planning the Dallas–Fort Worth region. We explore both, machine learning modeling approaches (e.g., ridge regression and random forest) and traditional statistical models (e.g., linear regression and multiplicative regression). This effort provides a detailed description of modeling workflows and of the preprocessing of input data including general transit feed specification (GTFS), employment, socio-demographic, and ridership data. We also describe metrics to compare model performance; in our experiments the ridge regression framework using a Yeo-Johnson power transformation led to the most accurate predictions with an [Formula: see text] of 0.88. The sensitivity of the DRM model to errors in the service-related predictor variables is within acceptable limits with the root mean squared error (RMSE) increasing by less than 20% for a 25% error in any one of the input predictors. Our findings suggest that DRMs can be a powerful complement to the four-step planning process, providing an alternative that is easier to maintain and run, and which may lead to more accurate ridership estimates given the limitations of transit modeling in traditional regional models. To illustrate the benefits of DRMs, this effort describes the deployment of trained models using a web-based framework which allows practitioners to obtain ridership estimates by drawing prospective routes on a map and providing a small number of service attributes as input.